Association of the 323/A3 Surface Glycoprotein with Tumor Characteristics and Behavior in Human Breast Cancer1

(CANCER RESEARCH 50. 3317-3321. June 1. 1990] Association of the 323/A3 Surface Glycoprotein with Tumor Characteristics and Behavior in Human Breast Cancer1

Atul K. Tandon, Gary M. Clark, Gary C. Chamness, and William L. McGuire2

Department of Medicine/Oncology, University of Texas Health Science Center at San Antonio, San Antonio, Texas 78284-7884

ABSTRACT antigen appears to be correlated with progression of the disease. The main focus of the present study was to determine more We have earlier described a monoclonal antibody (323/A3) against a fully the relationship between occurrence of 323/A3 protein M, 43,000 surface glycoprotein of MCF-7 human breast cancer cells with other variables as well as with clinical behavior in breast which shows considerable specificity for primary and metastatic breast tumors (Cancer Res., 46: 1306-1317, 1986). Here we report the occur cancer. rence of the 323/A3 antigen in a large cohort of primary breast tumors (n = 384) and its interrelationship with several clinically important MATERIALS AND METHODS variables. Frozen, stored tumor tissues were examined by a Western blot proce Materials and Reagents. Bicinchoninic acid (BCA) reagents (A and dure, and the level of 323/A3 protein in individual tumors was calculated B) for protein determination were purchased from the Pierce Chemical in arbitrary units based on the integrated M, 43,000 signal in tumors Co., Rockford, IL. Reagents for electrophoresis were obtained from compared with an MCF-7 internal standard. Thirty-six % (139 of 384) Bio-Rad Laboratories, Richmond, CA. Nitrocellulose membranes (0.45 of tumors were found to be positive for 323/A3. Higher frequencies of Mm) were purchased from Schleicher and Schuell, Keene, NH. '"I- 323/A3 protein were found in tumors larger than 2 cm (P = 0.03), tumors Labeled rabbit anti-mouse IgG was obtained from New England Nu with infiltrated lymph nodes (/' = 0.01), and tumors without estrogen clear, Boston, MA. All other chemicals used were of analytical purity receptor (P = 0.006). No significant relationship was found with patient grade or the best grade available. age, menopausa! status, or progesterone receptor status. Of the newer Acquisition and Storage of Human Breast Tumors. Tumor specimens clinical determinants proliferative rate (% S phase), DNA ploidy, and studied in this report were obtained from women undergoing surgery the lysosomal protease cathepsin D, but not the HER-2/neu oncogene for primary breast cancer in hospitals nationwide. Immediately after protein, were significantly correlated with 323/A3. The presence of 323/ surgical removal, a portion of each tumor specimen was snap frozen in A3 protein was also related to increased recurrence (P = 0.003) and liquid nitrogen and shipped to our laboratory on dry ice for steroid mortality (P = 0.036) after primary treatment. receptor analysis. All tissues were coded, mechanically pulverized in As an exposed surface antigen, this glycoprotein might be a useful the frozen state, and analyzed for receptors. The remaining powders target in radioimaging and immunotherapy of some human breast tumors, were stored in air-tight plastic tubes at -70Â°C.Tissues for the quanti- especially those having large size, infiltrated lymph nodes, deficient tation of 323/A3 antigen were obtained retrospectively from these estrogen receptor, high proliferative rate, abnormal DNA content, and biopsy specimens collected for clinical purposes unrelated to this inves high levels of cathepsin D, all of which are ominous indicators of tumor tigation. Sections 4 pm thick from formalin-fixed, paraffin-embedded behavior. tissues were stained with hematoxylin-eosin and examined under a microscope for the presence of tumor cells. All 384 specimens included in this study contained tumor cells, although the percentage of tumor INTRODUCTION cells varied considerably. Clinical Characteristics of Subjects and Follow-up. The clinical infor Although a number of tumor-associated cell surface antigens mation was abstracted from the individual patient chart. Tumors ana have been reported in breast cancer and have even been used in lyzed here were from a patient population ranging in age from 27 to therapeutic and imaging trials (1-12), the relationship of these 86 years with a median of 58 years. Lymph nodes were examined antigens with other known clinical determinants has not usually microscopically for the presence of tumor cells at the individual hos been examined. We have previously described a MAb,3 323/ pitals where surgery was performed. Lymph node involvement was A3, against a M, 43,000 surface glycoprotein of MCF-7 human considered negative when no lymph nodes contained tumor cells and positive if one or more nodes showed the presence of malignant cells. breast cancer cells (1). Based on a comparison by molecular Fifty-one % (195 of 384) of patients included in this investigation did weight and immunocytochemical tissue distribution, this M, not contain tumor extensions to the axillary lymph nodes. Of the node- 43,000 glycoprotein represents a protein previously undescribed positive patients, 62% (118 of 189) had more than 3 lymph nodes with either in breast cancer or in other tumors. Strong 323/A3 tumor infiltration. Tumor size (largest diameter) was recorded at the binding was observed to three human breast cancer cell lines time of surgery. Depending upon the size, tumors were divided into (MCF-7, T47D, and ZR-75), whereas two other human breast two groups for the purpose of statistical analysis. Of the 384 tumor cancer cell lines (MDA-231 and HBL-100) and three non-breast specimens studied in this report, 103 (27%) were less than or equal to cell lines displayed little or no binding (13). In histochemical 2 cm Â¡ndiameter. Levels of ER and PgR in tumor cytosols are routinely studies, this antibody detected 75% of metastatic lymph nodes determined in our laboratory as described previously (14, 15). Tumors and 59% of primary breast tumors and showed some staining were designated positive if the receptor binding was >3 fmol for ER and >5 fmol for PgR per mg cytosol protein. in 20% of benign breast lesions (1), so that presence of the Proliferative rate (percentage of S phase) and ploidy (DNA content) Received 10/10/89; revised 2/23/90. were determined in our laboratory by flow cytometric analysis (16). The costs of publication of this article Â«eredefrayed in part by the payment Levels of the HER-2/ni>Â«oncogene protein and the M, 34,000 mature of page charges. This article must therefore be hereby marked advertisement in form of cathepsin D were quantitated by Western blotting and densi- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by NIH Grant CA 30195. It was presented at the tometry of pertinent bands on the resulting autoradiograms as detailed 11th Annual San Antonio Breast Cancer Symposium. November 1988. before (17,18). Patients were followed for disease recurrence and overall 2Clinical Research Professor of the American Cancer Society. To whom survival as described previously (19). Median clinical follow-up time requests for reprints should be addressed, at University of Texas Health Science for patients still living was 65 months with a range of 16-168 months. Center. 7703 Floyd Curl Drive, San Antonio. TX 78284-7884. 3The abbreviations used are: MAb. monoclonal antibody; ER, estrogen recep Any postsurgical appearance of malignancy either near to or distant tor; PgR. progesterone receptor; 323/A3. a M, 43.000 surface glycoprotein first from the operated breast was considered as recurrence of disease. A identified in MCF-7 cells; SDS. sodium dodecyl sulfate. computer database containing continually updated clinical information 3317

on each patient was used for statistical analysis. MCF-7 Cells for 323/A3 Internal Standard. MCF-7 human breast cancer cells were cultured in Eagle's minimum essential medium sup plemented with 10 m.M4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid, 1% nonessential amino acids (GIBCO), 2 HIM L-glutamine (GIBCO), 25 Mg/ml gentamicin (Irvine Scientific), 6 ng/ml bovine insulin, and 5% calf serum (K. C. Biologicals). Sodium bicarbonate A560 (0.2%) was added to adjust the final pH to approximately 7.2. Cells were allowed to grow at 37Â°Cin an atmosphere containing 5% CO2. Logarithmically growing MCF-7 cells close to confluency were har vested by a brief incubation with 1 mM EDTA in phosphate-buffered saline. Cells were washed twice with phosphate-buffered saline, and the cell pellets were stored at -70Â°Cuntil required for assay. Western 43KDQ Quantitation of 323/A3 Protein. Proteins were extracted from the Blot tumor tissues or cell pellets as described elsewhere (17). In brief, approximately 10 mg of powder from each tumor were exposed to 150 1234 567 8,.100 50 25, n\ of 5% SDS. Tubes were kept in a boiling water bath for 5 min. Clear Human Breast Tumors MCF-7Stds supernatant was collected after spinning the tubes at 13,000 x g for 2 (Units 323/A3) min. Residual insoluble material was once again extracted with 150 ^1 Fig. I. Solimi nani nal inn of 323/A3 protein in breasl tumors by Western blot of 5% SDS as described above, and the first and second supernatants and densitometry. The 323/A3 band at M, 43.000 (kDa) along with corresponding were combined. Protein concentration in the SDS extracts was deter densitometric scanning is shown. The 323/A3 content in specimens was calculated in arbitrary units based on an internal standard of human breast cancer MCF-7 mined by the BCA method (20). cells. The 323/A3 values for the 8 tumors shown here are 113, 3, 4, 0, 0. 3. 28, Tumor proteins (200 ^g) were separated on 10% polyacrylamide and 3 units, respectively. vertical slab gels using 3% stacking gels. Electrophoresis was carried out under denaturing, non-reducing conditions (21). A SDS extract of MCF-7 human breast cancer cells was included at three concentrations (200, 100, and 50 /jg protein, corresponding to 100, 50, and 25 arbitrary units of 323/A3) in each gel as our laboratory standard. Transfer of resolved proteins onto nitrocellulose membranes was performed at 200 mAmp for 16 h at 4Â°C(22). Following blocking of nonspecific sites with 5% non-fat dry milk (1 h), the blots were incubated with 323/A3 monoclonal antibody (1 /jg/ml) overnight at 4Â°C.'"I-labeled rabbit anti-mouse IgG (100,000 cpm/ml) was used as the second antibody. CD After washing, the blots were exposed for 20-24 h to X-OMAT X-ray film (Kodak) at â€”70Â°Cusingintensifying screens. The level of 323/A3 CJ protein in individual tumors was quantitated by densitometric scanning of the M, 43,000 band on the resulting autoradiograms in a Beckman DU-7 spectrophotometer and was expressed in arbitrary units by com parison with an MCF-7 internal standard. Statistical Determinations. Statistical analysis of data for association between 323/A3 protein and other characteristics in breast cancer was 10 20 30 40 50 60 70 performed using 2-way contingency tables and nonparametric correla Median= 1 tion coefficients. The patient characteristics included were patient age at diagnosis, menopausa! status, tumor size, estrogen and progesterone 323/A3 Units receptor status, number of infiltrated lymph nodes, proliferative rate Fig. 2. Distribution of 323/A3 protein content in 384 human breast tumors. (percentage of S phase), DNA ploidy. HER-2/new, and cathepsin D. All computations were done with the BiomÃ©dicalComputer Programs- 384) of the tumors exhibited >3 units of 323/A3 protein. P Series (23). Survival curves were constructed by the method of Kaplan and Meier (24). The log-rank test for censored survival data was used Detection of low levels (1-2 units) of a protein by a semiquan to test the statistical significance of difference between the curves (25). titative Western blot procedure as used here is mainly a function of time of exposure of the nitrocellulose filters to the X-ray film. A tumor which after a given period of exposure appears RESULTS negative (0 unit) on the autoradiogram may appear as a faint band on a relatively longer exposure and is then calculated to 323/A3 Protein in Breast Tumors. We measured the level of be a low value (1-2 units) rather than zero. For these reasons, 323/A3 protein in 384 primary breast tumors using the semi- we considered >3 units of 323/A3 protein to be necessary for quantitative Western blot procedure described above. Fig. 1 classifying a tumor as 323/A3"1".In Fig. 1, three tumors (tumors shows a typical Western blot of 8 tumors and a corresponding 2, 6, and 8) with 3 units of 323/A3 are shown to appreciate densitometric scan of the 323/A3 bands at M, 43,000. In each this level of protein. gel 3 concentrations of the same extract of human breast cancer Stability of 323/A3 Protein. Because, prior to analysis for cells (MCF-7) were included as an internal laboratory standard. 323/A3 protein, the tumor specimens were stored at -70Â°C for 323/A3 units for the 8 tumors shown in Fig. 1 are given in the extended periods of time, it was important to find out the stability of this protein. The frequency of 323/A3"1"tumors was figure legend. Distribution of 323/A3 Protein. The level of 323/A3 protein found to be constant irrespective of the length of storage at ranged from 0 to 113 units with a median value of 1 unit. Its â€”70Â°C;asimilar proportion of tumors stored for less than 5 distribution in this group of patients is shown in Fig. 2. No years, 5-10 years, or more than 10 years showed the presence 323/A3 protein was detected in 33% ( 125 of 384) of the primary of 323/A3 protein (Table 1). The incidence of 323/A3 positivity breast tumors assessed. Only very low levels (1-2 units) were in the three groups was not statistically different (P = 0.61). (120 of 384) of the cases. Thirty-six % (139 of 323/A3 Protein versus Clinical Characteristics. After quanti- 3318

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1990 American Association for Cancer Research. 323/A3 GLVCOPROTEIN IN BREAST CANCER Table 1 Stability O/323/A3 protein in breast tumor specimens stored at â€”70Â°C patient, or menopausal status, although the trend for PgR Storage lime % 323/A3-positiveÂ° paralleled the difference for ER as expected. (yr) n Recently, a number of additional parameters have been re <5 26 38 ported in the literature to be related to clinical behavior of 5-10 264 38 94 32 breast cancer. These include flow cytometric determination of 0.61. the proliferativi rate of the tumor (percentage of S phase) and ploidy (DNA content) (16, 19), and measurement of the HER- Table 1 Relationship ofÃ¬2Ã¬jAÃŒproleinwith clinical characteristics in breast 2/neu oncogene protein (a growth factor receptor-like trans- cancer membrane glycoprotein with a molecular weight of 185,000- CharacteristicInfiltrated 323/A3-positive293946473227 190,000) (17) and cathepsin D (an estrogen-induced lyosomal nodes01-3>3Estrogen protease) (18). Values for all of these parameters were available on the tumor specimens analyzed in this report and their relationships with 323/A3 protein are also shown in Table 2. Every parameter receptor<3 fmol/mg protein associated with tumor aggressiveness (high S phase, aneuploidy, proteinTumor>3 fmol/mg high HER-2/neu and cathepsin D) was also associated with a greater incidence of 323/A3 positivity. Only the association size<2cm with HER-2/neu failed to reach statistical significance. >2cmProgesterone 281193 4040 323/A3 Protein and Clinical Behavior. We also examined directly the association of 323/A3 positivity with the likelihood receptor<5 fmol/mg protein of breast cancer recurrence and mortality. Kaplan-Meier life proteinPatient>5 fmol/mg 191122 3335373440 table analyses of both recurrence (Fig. 3) and mortality (Fig. 4) age<50 show that patients whose tumors have 323/A3 protein are vr significantly more likely to suffer disease recurrence and death, Â£50yrMenopausa! 2621041025812121178135 as expected for the associations with other characteristics al statusPrePeri ready presented. These survival results are summarized in Table 3. When all characteristics are included in multivariate analyses, however, 323/A3 protein does not emerge as an independent PostUnknownci 3825214825 predictor of recurrence or mortality (not shown). SphaseÂ£6.7 1 >6.7PloidyDiploid 0.9

_Q AneuploidHER-2/neu 216313 4235 CD 0.8 _Q O protein<100 Â£ 0.7 323/A3- units CD > 100unitsCathepsin 69229153% 443143P0.010.0060.030.130.790.79<0.00010.0010.160.02.> 06 n = 245

Dprotein<75 05 units275 5 p = 0.003 unitsÂ«19571118111273103 0.4- 323/A3+ n = 139 tation of 323/A3 protein content in specimens, tumors were 0 12 24 36 48 60 72 84 decoded for statistical correlations. The relationships of this Time (months) protein to several known clinical characteristics of the patients Fig. 3. Time to recurrence in a total of 384 breast cancer patients with 323/ and their breast tumors are shown in Table 2. A3" or 323/A3* tumors. The recurrence curves are based on a clinical follow-up An important clinical characteristic of breast cancer patients period of 16 to 168 months, with a median of 65 months for living patients. is the presence of metastatic tumor deposits in the axillary lymph nodes at the time of primary surgery. We had already shown that the 323/A3 protein is more prevalent in nodal mÃ©tastasesthan in primary tumors (1). It was of interest, therefore, to determine the relationship between nodal status and presence of 323/A3 protein in the primary tumors. Of the 384 patients in this study, 51% (195 of 384), 18% (71 of 384), and 31% (118 of 384) had 0, 1-3, and >3 axillary lymph nodes infiltrated with tumor cells, respectively (Table 2). Tumors with many infiltrated nodes contained a substantially higher propor tion of 323/A3+ breast cancers (P = 0.01). Tumor size and steroid receptor deficiency are also known to be associated with clinical aggressiveness (26). We found a â€¢n greater frequency of 323/A3 positivity in tumors deficient in 12 24 36 48 60 72 84 estrogen receptor (P = 0.006), and in tumors >2 cm in size (P Time (months) = 0.03). No significant correlation was found between the 323/ Fig. 4. Survival probabilities in 384 breast cancer patients in relation to 323/ A3 positivity of the primary tumor and PgR status, age of the A3 status. Other details are same as given in the legend to Fig. 3. 3319

Table 3 Five-year actuarial recurrence and mortality of 384 breast cancer ness, tumor cell proliferation, and abnormal DNA content. patients as a function of 323/A3 status Thus, it is not surprising that the 323/A3 protein is related to reduced disease-free and overall survival in univariate analyses %Â±SE52 %Â±SE39 (Figs. 3 and 4). In multivariate settings, however, 323/A3 is Recurrence Â±4 Â±323 Mortality323/A3-positive34 Â±4323/A3-negative Â±3P"0.003 0.036 not an independent factor of disease recurrence or overall Â°Log-rank test. survival (data not shown). Nevertheless, in light of the fact that the 323/A3 glycoprotein DISCUSSION is an exposed surface antigen and is frequently present on tumors with poor prognosis, it may offer a potential clinical The availability of MAbs reacting specifically with human target for immunodetection and/or immunotherapy of such breast tumor-associated cell surface antigens may offer an op breast cancers. In fact, there is preliminary evidence for both of portunity to diagnose and treat cancers more effectively. The these applications of this MAb. (a) Radiolabeled MAb 323/A3 MAb 323/A3, used in the present study to quantitate an antigen has been successfully used to image progressively growing BT- in breast tumors, is an IgGl generated by immunizing BALB/ 20 human breast tumors in athymic mice (27). Of the several c mice with the human breast tumor MCF-7 cell line. This labels and IgG fragments examined, 323/A3 Fab' labeled with 99MTcwas suggested to have the highest potential clinical di MAb shows specificity for primary and metastatic human breast tumors by immunoperoxidase staining in histochemistry (1). It agnostic utility since it allowed earlier detection of tumors, recognizes a M, 43,000 membrane-associated antigen in cell better tumortissue tracer ratios, and a more rapid blood clear lines as well as in breast tumors. This antigen is identified as a ance (28). (b) A potent immunotoxin has been prepared by glycoprotein based on its ability to bind to and specifically elute linking the 323/A3 MAb with the toxic A chain of ricin. The from a concanavalin A column. Moreover, the MAb precipitates cytotoxicity of this conjugate in cultured cell lines depends a labeled M, 43,000 protein from MCF-7 cells after pulse upon the level of 323/A3 antigen in each line (13). labeling with ['Hjglucosamine (1). All of these preclinical studies suggest that MAb 323/A3 In the present study, quantitation of 323/A3 antigen in breast may prove to be a useful diagnostic and therapeutic tool in tumors was done by Western blot and densitometry of the breast cancer. resulting M, 43,000 bands on autoradiograms. Protein samples for 323/A3 Western blotting were electrophoresed under non- ACKNOWLEDGMENTS reducing conditions (absence of ÃŸ-mercaptoethanol or dithio- threitol) because the 323/A3 antigenic site is sensitive to reduc The authors wish to thank Mary Nell Baird and Robert Duarte for ing agents. their skillful technical assistance with the 323/A3 assays, Judy Wenzel Results presented here show that 323/A3 protein is fre for data management, David Mascorro for his computer programming, quently present in human breast tumors. Thirty-six % (139 of and Sandy Montgomery for her patience in typing this manuscript. 384) of tumors assessed appeared clearly positive for 323/A3. Occurrence of this protein was analyzed for association with REFERENCES several other characteristics of patients and their breast tumors. 1. Edwards, D. P., Grzyb, K. T.. Dressler, L. G., Mansel, R. E., Zava, D. T., A greater frequency of 323/A3 protein was seen in patients Sledge. G. W., Jr.. and McGuire, W. L. Monoclonal antibody identification with tumor-infiltrated lymph nodes, suggesting that the colo and characterization of a M, 43,000 membrane glycoprotein associated with nization of lymph nodes with tumor cells may be affected by human breast cancer. Cancer Res., 46: 1306-1317. 1986. 2. Nuti, M., Teramoto, Y. A., Mariani-Costantini. R.. Horan-Hand. P.. the presence of this protein on their surface. Positivity for 323/ Colcher, D., and Schlom, J. A monoclonal antibody (B72.3) defines patterns A3 protein was also related to tumor burden taken as the size of distribution of a novel tumor-associated antigen in human mammary of the excised primary tumor; larger tumors (>2 cm in diameter) carcinoma cell populations. Int. J. Cancer. 29: 539-545, 1982. 3. Yuan, D., Hendler, F. J., and Vitella, E. S. Characterization of a monoclonal were more often positive for 323/A3 protein (40% versus 27%). antibody reactive with a subset of human breast tumors. J. Nati. Cancer Inst., Because tumor size may be in part a reflection of the prolifer- 68:119-728. 1982. 4. Ceriani, R. L.. Sasaki, M., Sussman, H., Wara, W. M., and Blank, E. W. ative rate of the tumor, one might also expect a correlation Circulating human mammary epithelium antigens in breast cancer. Proc. with percentage of S phase. Indeed, a highly significant (P < Nati. Acad. Sci. USA, 79: 5420-5424. 1982. 0.0001) relationship was found. A strong direct correlation of 5. Soule, H. R., Linder, E., and Edgington. T. S. Membrane 126-kilodalton phosphoglycoprotein associated with human carcinomas identified by a hy- 323/A3 protein expression with nodal infiltration, tumor size, bridoma antibody to mammary carcinoma cells. Proc. Nati. Acad. Sci. USA, and proliferation rate suggests that 323/A3 may be related to 80: 1332-1336, 1983. 6. Croghan, G. A., Papsidero, L. D., Valenzuela, L. A., Nemoto, T., Penetrante, tumor aggressiveness and/or its metastatic behavior. Presence R., and Chu, T. M. Tissue distribution of an epithelial and tumor-associated of 323/A3 protein was also significantly associated with absence antigen recognized by monoclonal antibody F36/22. Cancer Res., 43: 4980- of ER and thus presumably with loss of estrogen responsiveness 4988. 1983. 7. Menard. S.. Tagliabue, E., Canevari, S., Fossati, G., and Colnaghi, M. I. in tumors. Generation of monoclonal antibodies reacting with normal and cancer cells No significant correlation was found in the 323/A3 and HER- of human breast. Cancer Res., 43: 1295-1300, 1983. 2/neu oncogene protein in these primary tumors, although we 8. Thompson. C. H., Jones, S. L., Whitehead. R. H., and McKenzie, I. F. C. A human breast tissue-associated antigen detected by a monoclonal antibody. have reported a prognostic significance of the HER-2/neu pro J. Nati. Cancer Inst., 70: 409-419, 1983. tein in node-positive breast cancer patients (17). A significant 9. Burchell, J., Wang, D., andTaylor-Papadimitriou. J. Detection of the tumor- (P = 0.02) relationship did appear, however, between the 323/ associated antigens recognized by the monoclonal antibodies HMFG-1 and -2 in serum from patients with breast cancer. Int. J. Cancer, 34: 763-768, A3 protein and the M, 34,000 mature form of cathepsin D, a 1984. strong independent parameter of disease recurrence and mor 10. Hayes, D. F., Sekine. H., Ohno. T., Abe, M., Keefe, K., and Kufe, D. W. Use of a murine monoclonal antibody for detection of circulating plasma tality in node-negative breast cancer (18). DF3 antigen levels in breast cancer patients. J. Clin. Invest., 75:1671-1678, Although the direct biological role of 323/A3 protein in 1985. breast tumor growth is unclear, these findings suggest that the 11. White, C. A., Dulbecco. R.. Allen, R., Bowman, M., and Armstrong, B. Two monoclonal antibodies selective for human mammary carcinoma. Cancer occurrence of this protein in tumors is in part associated with Res.. 45: 1337-1343. 1985. metastatic progression of disease, loss of estrogen responsive 12. Hilkens. J.. Kroezen, V., Bonfrer, J. M. G., De Jong-Bakker. M.. and 3320

Bruning, P. F. MAM-6 antigen, a new serum marker for breast cancer 20. Smith. P. K.. Krohn. R. I.. Hermanson, G. T., Mallia, A. K.. Gartner. F. H., monitoring. Cancer Res., 46: 2582-2587. 1986. Provenzano. M. D., Fujimoto, E. K., Goeke, N. M., Olson. B. J., and Klenk, 13. LeMaistre, C. F., Edwards, D. P., Krolick, K. A., and McGuire, W. L. An D. C. Measurement of protein using bicinchoninic acid. Anal. Biochem., 750: immunotoxin cytotoxic for breast cancer cells in vitro. Cancer Res.. 47: 730- 76-85, 1985. 734, 1987. 21. Laemmli, U. K. Cleavage of structural proteins during the assembly of the 14. McGuire, W. L., De La Garza, M., and Chamness, G. C. Evaluation of head of bacteriophage T4. Nature (Lond.). 227: 680-685. 1970. estrogen receptor assays in human breast cancer tissue. Cancer Res., 3 7:637- 22. Towbin. H.. Staehelin. T., and Gordon. J. Electrophoretic transfer of proteins 639. 1977. from polyacrylamide gels to nitrocellulose sheets: procedures and some 15. Powell, B.. Carola. R. E.. Chamness. G. C., and McGuire. W. L. Measure applications. Proc. Nati. Acad. Sci. USA. 76:4350-4354. 1979. ment of progesterone receptor in human breast cancer biopsies. Cancer Res., 23. Dixon. W. J. BMDP Statistical Software. Berkeley, CA: University of Cali 39: 1678-1682. 1979. fornia Press. 1985. 16. Dressier, L. G.. Seamer, L. C., Owens, M. A., Clark, G. M., and McGuire, 24. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations. J. Am. Stat. Assoc., 53: 457-481, 1958. W. L. DNA flow cytometry and prognostic factors in 1331 frozen breast cancer specimens. Cancer (Phila.). 61: 420-427. 1988. 25. Mantel. N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother. Rep.. 50: 163-170. 1966. 17. Tandon. A. K.. Clark, G. M.. Chamness, G. C., Ullrich, A., and McGuire, 26. Clark. G. M.. and McGuire. W. L. Steroid receptors and other prognostic W. L. HER-2/WU oncogene protein and prognosis in breast cancer. J. Clin. factors in primary breast cancer. Semin. Oncol.. /5: 20-25, 1988. Oncol., 7: 1120-1128, 1989. 27. Khaw, B. A., Bailes, J. S.. Schneider, S. L., Lancaster, J., Powers, J., Strauss, 18. Tandon. A. K.. Clark. G. M.. Chamness, G. C., Chirgwin, J. M., and H. W., Lasher, J. C., and McGuire, W. L. Human breast tumor imaging McGuire. W. L. Cathepsin D and prognosis in breast cancer. N. Engl. J. using '"In-labeled monoclonal antibody: athymic mouse model. Eur. J. NucÃ. Med.. 322: 297-302. 1990. Med.. 14: 362-366. 1988. 19. Clark. G. M., Dressier, L. G., Owens, M. A., Pounds. G.. Oldaker, T., and 28. Pak, K. Y., Nedelman, M. A., Fogler, W. E., McGuire, W. L., and Daddona, McGuire, W. L. Prediction of relapse or survival in patients with node- P. E. Rapid detection of human breast carcinoma xenografts using techne- negative breast cancer by DNA flow cytometry. N. Engl. J. Med., 320: 627- tium-99M 323/A3 Fab' in an athymic nude mouse model. Breast Cancer 633. 1989. Res. Treat., 12: 145, 1988.

3321

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1990 American Association for Cancer Research. Association of the 323/A3 Surface Glycoprotein with Tumor Characteristics and Behavior in Human Breast Cancer

Atul K. Tandon, Gary M. Clark, Gary C. Chamness, et al.

Cancer Res 1990;50:3317-3321.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/50/11/3317

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/50/11/3317. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.